Maximal Characteristic Sets and Neighborhoods Approach to Incomplete Information Systems

  • Chien-Chung Chan
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7413)


Data with missing values are represented as incomplete information systems in rough sets approaches. There are two possible interpretations of missing values as “do not care” or “lost” values. Many existing works considered only the former case. The use of characteristic sets to deal with both cases was first introduced by Grzymala-Busse. In this paper, we introduce a refinement of characteristic set approach to incomplete information systems, and we show that it can improve approximation accuracy similar to the improvement obtained by applying the techniques of maximal consistent blocks and binary neighborhood systems approaches to dealing with “do not care” missing values. Additionally, subset and concept based approximations are introduced for binary neighborhood systems in order to preserve upper approximations.


Rough sets Incomplete information systems Characteristic sets Maximal characteristic sets Binary neighborhood systems 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Pawlak, Z.: Rough sets: basic notion. International Journal of Computer and Infor-mation Science 11(15), 344–356 (1982)Google Scholar
  2. 2.
    Pawlak, Z.: Rough Sets and Decision Tables. In: Skowron, A. (ed.) SCT 1984. LNCS, vol. 208, pp. 186–196. Springer, Heidelberg (1985)Google Scholar
  3. 3.
    Pawlak, Z.: Rough Sets: Theoretical Aspects of Reasoning about Data. Kluwer Academic Publishing (1991)Google Scholar
  4. 4.
    Kryszkiewicz, M.: Rough set approach to incomplete information systems. Information Sciences 112, 39–49 (1998)MathSciNetzbMATHCrossRefGoogle Scholar
  5. 5.
    Leung, Y., Li, D.: Maximal consistent block techniques for rule acquisition in incomplete information systems. Information Sciences 153, 85–106 (2003)MathSciNetzbMATHCrossRefGoogle Scholar
  6. 6.
    Grzymała-Busse, J.W.: Characteristic Relations for Incomplete Data: A Generalization of the Indiscernibility Relation. In: Tsumoto, S., Słowiński, R., Komorowski, J., Grzymała-Busse, J.W. (eds.) RSCTC 2004. LNCS (LNAI), vol. 3066, pp. 244–253. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  7. 7.
    Grzymala-Busse, J.W.: On the Unknown Attribute Values in Learning from Examples. In: Raś, Z.W., Zemankova, M. (eds.) ISMIS 1991. LNCS (LNAI), vol. 542, pp. 368–377. Springer, Heidelberg (1991)CrossRefGoogle Scholar
  8. 8.
    Grzymała-Busse, J.W., Rząsa, W.: Local and Global Approximations for Incomplete Data. In: Peters, J.F., Skowron, A. (eds.) Transactions on Rough Sets VIII. LNCS, vol. 5084, pp. 21–34. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  9. 9.
    Stefanowski, J., Tsoukiàs, A.: On the Extension of Rough Sets under Incomplete Information. In: Zhong, N., Skowron, A., Ohsuga, S. (eds.) RSFDGrC 1999. LNCS (LNAI), vol. 1711, pp. 73–82. Springer, Heidelberg (1999)CrossRefGoogle Scholar
  10. 10.
    Yang, X., Zhang, M., Dou, H., Yang, J.: Neighborhood systems-based rough sets in in-complete information system. Knowledge-Based Systems 24, 858–867 (2011)CrossRefGoogle Scholar
  11. 11.
    Yao, Y.Y.: Two views of the theory of rough sets in finite universes. International J. of Approximate Reasoning 15, 291–317 (1996)zbMATHCrossRefGoogle Scholar
  12. 12.
    Lin, T.Y.: Neighborhood systems and approximation in database and knowledge base systems. In: Proceedings of the Fourth International Symposium on Methodologies of Intelligent Systems, Poster Session, October 12-15, pp. 75–86 (1989)Google Scholar
  13. 13.
    Lin, T.Y.: Granular computing I: the concept of granulation and its formal model. International Journal of Granular Computing, Rough Sets and Intelligent Systems 1, 21–42 (2009)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Chien-Chung Chan
    • 1
  1. 1.Department of Computer ScienceUniversity of AkronAkronUSA

Personalised recommendations